1 2+++ This is experimental code +++ 3 4Authors: Brian Nielsen, Michael Schwendt, �ge R�bekk, Esa Pulkkinen, 5 Phillip Wooller, and contributors. 6 7Porting hints: The longer the sample buffer, the more time the CPU spends 8on consecutively pre-calculating new sample data without feeding the audio 9driver. In case of interruptions due to heavy CPU usage the audio driver 10might run out of data. Filling and sending shorter buffers to the audio 11driver is more likely to keep up a constant data stream. Additionally, the 12OSS/Free driver for Linux provides methods to split its internal audio buffer 13into fragments, which get send to the soundcard separately, but one after each 14other. Hereby the driver can quickly transfer new data to the soundcard, 15whenever the card finished processing previous data. Depending on the 16available CPU power it is recommended to find a mean buffer size. Currently, 17the buffer size is determined by the two (originally OSS/Free-specific) 18parameters ``fragments'' and ``fragSizeBase'', where: 19 bufferSize = fragments * 2^fragSizeBase 20 or without fragmentation: bufferSize = 2^fragSizeBase 21 22This does not necessarily apply to every audio device driver. 23 24The variables ``encoding, precision, channels'' are intended to contain 25SID-Emulator compatible audio configuration values, which can be read out 26to configure the SID-Emulator. 27